1

e

Protist, Vol. 163, 400–414, May 2012http://www.elsevier.de/protisPublished online date 11 August 2011

ORIGINAL PAPER

Sphagnum-dwelling Testate Amoebae in SubarcticBogs are More Sensitive to Soil Warming in theGrowing Season than in Winter: the Results ofEight-year Field Climate Manipulations

Andrey N. Tsyganova,1, Rien Aertsb, Ivan Nijsc,d, Johannes H.C. Cornelissenb, andLouis Beyensa

aResearch group Polar Ecology, Limnology and Geomorphology, Department of Biology, University ofAntwerp, Universiteitsplein 1, B-2610 Antwerp (Wilrijk), Belgium

bSystems Ecology, Department of Ecological Science, VU University, De Boelelaan 1085, NL-1081 HVAmsterdam, the Netherlands

cResearch group Plant and Vegetation Ecology, Department of Biology, University of Antwerp,Universiteitsplein 1, B-2610 Antwerp (Wilrijk), Belgium

dKing Saud University, Riyadh, Saudi Arabia

Submitted April 4, 2011; Accepted July 4, 2011Monitoring Editor: Bland J. Finlay

Sphagnum-dwelling testate amoebae are widely used in paleoclimate reconstructions as a proxy forclimate-induced changes in bogs. However, the sensitivity of proxies to seasonal climate componentsis an important issue when interpreting proxy records. Here, we studied the effects of summer warm-ing, winter snow addition solely and winter snow addition together with spring warming on testateamoeba assemblages after eight years of experimental field climate manipulations. All manipulationswere accomplished using open top chambers in a dry blanket bog located in the sub-Arctic (Abisko,Sweden). We estimated sensitivity of abundance, diversity and assemblage structure of living andempty shell assemblages of testate amoebae in the living and decaying layers of Sphagnum. Our resultsshow that, in a sub-arctic climate, testate amoebae are more sensitive to climate changes in the growingseason than in winter. Summer warming reduced species richness and shifted assemblage composi-tion towards predominance of xerophilous species for the living and empty shell assemblages in bothlayers. The higher soil temperatures during the growing season also decreased abundance of empty

shells in both layers hinting at a possible increase in their decomposition rates. Thus, although pos-sible effects of climate changes on preservation of empty shells should always be taken into account,species diversity and structure of testate amoeba assemblages in dry subarctic bogs are sensitiveproxies for climatic changes during the growing season.© 2011 Elsevier GmbH. All rights reserved.

Key words: Bogs; climate; open top chambers; Sphagnum; subarctic; testate amoebae.

Corresponding author; fax +32 3 265 29 50-mail andrey.tsyganov@ua.ac.be (A.N. Tsyganov).

Abbreviations: OTC, open top chambers.

© 2011 Elsevier GmbH. All rights reserved.doi:10.1016/j.protis.2011.07.005

Seasonal Climate Effects on Testacea 401

Introduction

Understanding of climate-ecosystem relationshipsis important for development of paleoclimate recon-struction techniques and estimation of possibleeffects of climate change. It has been shown thatclimate changes can considerably affect all lev-els of ecological organisation, including ranges ofspecies distribution, community composition andecosystem functioning (McCarty 2001; Norf et al.2007; Traill et al. 2010; Walther et al. 2002).Ecosystem responses to climate change will greatlydepend on the considered range on environmen-tal conditions because the rates of many biologicalprocesses have unimodal relationships with phys-ical parameters (Shaver et al. 2000). This meansthat effects of climate change will vary seasonallyand spatially being positive when environmen-tal parameters are shifted in the direction to theoptimum and negative otherwise. Thus, providingthe complexity of climate-ecosystem relationships,studies on seasonal and spatial variation of climaticeffects are necessary for reliable estimation of cli-mate impacts on ecosystems.

High-latitude bog ecosystems are known to bevery sensitive to climate conditions because theirfunctioning is primarily controlled by temperatureand precipitation (Moore and Bellamy 1974). Cli-mate models predict that, during this century,both average temperature and annual precipita-tion in the sub-Arctic will increase and the greatestincrease in those factors will take place in win-ter (Christensen et al. 2007). These changes canaffect all components of bog ecosystems includingdiverse and abundant assemblages of Sphagnum-dwelling testate amoebae. Testate amoebae areshelled protozoa which play an important role infood webs as decomposers of organic matter andas a trophic link between bacteria and invertebrates(Gilbert et al. 1998; Wilkinson and Mitchell 2010).They are valuable bioindicators and are widely usedfor paleoclimate reconstructions (Charman 2001;Mitchell et al. 2008; Tolonen 1986). Sphagnum-dwelling testate amoebae represent good modelorganisms for studying climate effects because oftheir small size and short-generation times whichallow testing hypotheses in an experimental set-up.In addition, testate amoebae have specific dis-tribution along Sphagnum stems in response tovertical gradients of light, temperature, moisture,oxygen, etc. (Mitchell and Gilbert 2004) so theycan be used for estimation of spatial variation of cli-mate effects. However, the relationships betweentestate amoeba assemblages in Sphagnum andclimate remain poorly understood and usually are

not differentiated among seasonal components ofclimate.

Changes in seasonal climate can affect bothliving and empty shell assemblages of Sphagnum-dwelling testate amoebae in the sub-Arctic.Increased temperatures during the growing seasoncan either increase or reduce abundance of livingorganisms depending on whether the temperatureshifts towards or away from the ecological optimum(Savage et al. 2004). Higher summer tempera-tures may have even stronger effects on abundancethrough changes in species diversity and assem-blage structure (Beveridge et al. 2010; Ives 1995;Jiang and Morin 2004; Petchey et al. 1999).In addition, summer warming may affect testateamoebae via decreasing temperature-dependentmoisture, especially in dry biotopes, because tes-tate amoebae in Sphagnum are known to bestrongly controlled by moisture availability in sum-mer (Lamentowicz and Mitchell 2005; Mitchell et al.1999). Increased winter precipitation can lead to athicker snow cover and, as a consequence, morefavourable temperatures for testate amoebae inSphagnum possibly resulting in a greater abun-dance. Besides, a thicker snow cover can alsoaffect environmental conditions at the beginningof the subsequent spring by delaying the onsetof the growing season or changing soil moisture(Lamentowicz et al. 2010; Maxwell 1992). However,higher spring temperatures can counteract such ashortening of the growing season and increase soilmoisture due to snow melt (Maxwell 1992). Thesechanges can positively affect testate amoeba abun-dance and shift assemblage composition towardspredominance of hydrophilous species as a resultof favourable moisture and temperature regimes.The effects of both increased winter precipitationalone and in combination with spring warming canbe long-lasting so that the effects can be detectedduring the subsequent growing season. Emptyshell assemblages can be also affected by thechanges in the climate parameters through shiftsin the living testate amoeba assemblages and/orby altering biological decomposition (Wilkinson andMitchell 2010).

So far, there has been a lack of in situ exper-imental studies on relationships between testateamoebae and climate. Most previous field climatemanipulation experiments were aimed at studyingresponses of soil testate amoeba assemblages towarming or increased precipitation separately, werefocused on changes in summer climate only and didnot last longer than one growing season (Beyenset al. 2009; Lousier 1974a, b; Tsyganov et al.2011). Although these experiments are important

402 A.N. Tsyganov et al.

for understanding of effects of a particular cli-mate parameter during the period of the highestactivity of testate amoebae, they fail in detect-ing climate effects in other seasons. Therefore,the aim of the present work is to investigateresponses of Sphagnum-dwelling testate amoe-bae to changes in seasonal climate componentsin a field manipulation experiment. Using open topchambers, we simulated summer warming, win-ter snow addition solely and winter snow additionin combination with spring warming in a relativelydry blanket bog located in sub-arctic Sweden (seeAerts et al. (2004) and Dorrepaal et al. (2004) forfurther details). The effects of the treatments weredetected on both living and empty shell assem-blages of testate amoebae in living (upper) anddecaying (deeper) layers of Sphagnum at the endof the growing season after eight years of the treat-ment. We focused on abundance, species diversityand assemblage structure of testate amoebae andhypothesized that: (i) Summer warming reducesabundance and diversity, mostly through decreas-ing temperature-dependent moisture content, andcauses shifts in assemblage structure towards pre-dominance of xerophilous species. (ii) Winter snowaddition increases abundance and diversity of tes-tate amoebae by mitigating negative effect of lowwinter temperatures. (iii) Winter snow addition incombination with spring warming enhances abun-dance and diversity of testate amoebae and shiftsassemblage structure towards predominance ofhydrophilous species. (iv) The effects of the climatemanipulations during various seasons are interac-tive and layer-specific.

Results

Composition of Testate AmoebaAssemblages

The analysis of the samples revealed 36 testateamoeba taxa belonging to 15 genera (Appendix A).Five species, Arcella gibbosa, Difflugiella oviformis,Hyalosphenia papilio, Nebela (Argynnia) dentis-toma, Nebela flabellulum, were only encounteredas empty shells, whereas all taxa from the livingassemblage also occurred as empty shells. Bothassemblages were dominated by Assulina mus-corum and Arcella catinus which comprised 49%and 54% of the total counts of living amoebaeand empty shells, respectively, and were observedin more than 80% of the samples. However, theassemblages differed in the composition of sub-dominant species. Bullinularia indica, Trigonopyxis

minuta, Corythion dubium were characteristic forthe living assemblage, whereas Nebela militaris,C. dubium, Archerella flavum were typical forthe empty shell assemblage. These species wereobserved in more than 50% of the samples andtogether with the two abovementioned dominanttaxa accounted for about 75% of the correspondingtotal counts.

Abundance of Testate AmoebaAssemblages

Concentrations of empty shells varied between thelayers and among the treatments whereas no sig-nificant differences were detected for abundanceof living amoebae (Table 1, Fig. 1A, B). Abun-dance of empty shells was greater in the deeperlayer (Table 1, Fig. 1B; F1,44 = 72.78, P < 0.001)and was decreased by the summer and win-ter climate manipulations in both layers (Table 1,Fig. 1B; F1,44 = 15.23, P < 0.001 and F2,44 = 6.68,P < 0.01, respectively). Averaged across the treat-ments, the mean concentrations of empty shellswere 24.5 ± 5.9 (SE) and 124.4 ± 19.2 * 103 shellsg-1 Sphagnum dry weight in the upper and deeperlayer, respectively. In both layers, the summerwarming treatment reduced the concentrations ofempty shells on average by 54%. Planned orthogo-nal comparisons for the winter treatments revealedthat only snow addition plus spring warming neg-atively affected concentrations of empty shells(F1,44 = 11.00, P = 0.002) reducing them by 58% inboth layers.

Diversity of Testate AmoebaAssemblages

Species diversity of the living testate amoebaassemblages, estimated with Shannon-Wiener’sdiversity index, increased with sampling depthand was reduced by the summer warming treat-ment (Table 1, Fig. 1C; F1,44 = 13.10, P = 0.001and F1,44 = 14.79, P < 0.001, respectively). Aver-aged across the treatments, mean values of thediversity index in the upper and deeper layerwere 1.46 ± 0.07 (SE) and 1.79 ± 0.07, respec-tively. The greater values of the diversity indexin the deeper layer was mostly related to moreequally distributed abundances among the species,because Pielou’s evenness index in that layer was0.84 ± 0.02 comparing to 0.66 ± 0.02 in the upperlayer (Table 1, Fig. 1G; F1,44 = 36.15, P = 0.001).Averaged across the treatments, species richnessof the living assemblage did not differ between thelayers. The summer warming treatment reduced

Seasonal Climate Effects on Testacea 403

Living Assemblage Empty Shell Assemblage

LayerLayer

Figure 1. Effects of the summer and winter climate manipulations on the univariate characteristics (A, B – totalconcentration; C, D – Shannon-Wiener’s diversity index; E, F – species richness; G, H – Pielou’s evennessindex) of the living and empty shell assemblages of testate amoebae in the upper (living) and deeper (decaying)layers of Sphagnum after eight years of the experimental treatments. See Table 2 for the treatment codes. Dataare means ± 1 SE (n = 5).

404 A.N. Tsyganov et al.

Table 1. The results (F-ratio and P-statistics) of the multifactorial ANOVA testing for the effects of the summerand winter climate manipulations on univariate characteristics of living and empty shell assemblages of testateamoebae in the living (upper) and decaying (deeper) Sphagnum layers after eight years of the treatments. Allparameters were loge transformed prior the analysis. +P < 0.1, *P < 0.05, **P < 0.01, ***P < 0.001.

Source of variation df Abundance Shannon’s diversity Species richness Pielou’s evenness

Living assemblageLayer 1 0.02 13.10*** 0.62 36.15***

Summer treatments 1 2.11 14.79*** 15.51*** 3.80+

Winter treatments 2 2.19 0.53 0.83 0.10Layer × Summer 1 0.81 0.04 0.00 0.02Layer × Winter 2 0.66 2.83+ 1.82 1.89Summer × Winter 2 1.83 2.24 0.12 3.84*

Layer × Summer × Winter 2 0.26 0.74 0.12 0.77Residuals 44

Empty shell assemblageLayer 1 72.78*** 30.22*** 26.28*** 10.33**

Summer treatments 1 15.23*** 10.63** 19.06*** 1.29Winter treatments 2 6.68** 2.14 2.70+ 1.03Layer × Summer 1 0.78 0.01 0.00 0.07Layer × Winter 2 0.53 2.97+ 0.94 1.64Summer × Winter 2 1.87 1.30 0.39 1.62Layer × Summer × Winter 2 0.20 0.13 0.54 0.12Residuals 44

species diversity of the living assemblage by about20% in both layers. In contrast to the vertical vari-ation in diversity, this trend was mostly driven by adecrease in species richness which was reducedon average by 27% (Table 1, Fig. 1E; F1,44 = 15.51,P < 0.001). The effect of the summer warming treat-ment on evenness was significant only in interactionwith the winter climate manipulations (Table 1;F2,44 = 3.84, P = 0.029), in the sense that evennessdecreased only in the absence of OTCs in winter(one-way ANOVA, F1,14 = 11.09, P = 0.005) and wasunaffected otherwise.

Species diversity of the empty shell assemblagesgenerally followed similar patterns, being enhancedby sampling depth and reduced by the summerwarming treatment (Table 1, Fig. 1D; F1,44 = 30.22,P < 0.001 and F1,44 = 10.63, P = 0.002, respec-tively). Averaged across the treatments, meanvalues of Shannon-Wiener’s diversity index were1.38 ± 0.06 (SE) and 1.85 ± 0.06 in the upper anddeeper layer, respectively. However, in contrast tothe living assemblage, the greater species diversityof empty shells in the deeper layer was associatedwith both greater species richness (15.0 ± 0.7 in thedeeper layer comparing to 10.8 ± 0.6 in the upperone; Table 1, Fig. 1D; F1,44 = 26.28, P < 0.001)and greater evenness (0.69 ± 0.02 in the deeperlayer comparing to 0.59 ± 0.02 in the upper one;

Table 1, Fig. 1H; F1,44 = 10.33, P = 0.002). The sum-mer warming treatment reduced species diversityof the empty shell assemblage on average by 14%in both layers. Similar to the living assemblages, thiswas mostly driven by changes in species richness,which was decreased on average by 23% in bothlayers (Table 1, Fig. 1D; F1,44 = 19.06, P < 0.001),whereas evenness was unaffected.

In both living and empty shell assemblages, mostof the species whose distribution was restricted toone layer only were rare. Usually they occurredin less than 10% of the samples and had meanrelative abundances below 7% (Appendix A). Theonly exception was Arcella arenaria which wasencountered only in the upper layer but had a liveoccurrence of 23%. However, empty shells of thespecies were found in both layers. Interestingly,all species of the genus Difflugia were observedin the deeper layer only. The more unequal dis-tribution of species abundances in the living andempty shell assemblages in the upper layer canbe accounted for by dominance of Assulina mus-corum. This species comprised 40% and 55%of the total counts of the living and empty shellassemblages in the upper layer, whereas otherspecies in that layer were considerably less abun-dant. On the other hand, the dominant species inthe deeper layer were less abundant compared with

Seasonal Climate Effects on Testacea 405

the corresponding assemblages in the upper layer(Trigonopyxis minuta comprised 15% of all livingindividuals and A. muscorum comprised 38% of allempty shells in the deeper layer). In addition, sub-dominant species in the deeper layer were morenumerous and abundant (Appendix A). In both lay-ers, the decreased species richness of the livingand empty shell assemblages in response to sum-mer warming was mostly related to the species withlow occurrence and relative abundance. Amongthe fifteen taxa that were not observed in theplots exposed to summer warming, two (Arcellahemisphaerica undulata, Euglypha rotunda) in theliving assemblage and only one (Euglypha filiferaspinosa) in the empty shell assemblage had occur-rences in the range between 10 and 50% of thesamples, but their relative abundance per samplestill never exceeded 12%.

Structure of Testate AmoebaAssemblages

Partial redundancy analyses showed that samplingdepth and summer warming explained a signifi-cant part of the total variation in the assemblagestructure of both living amoebae and empty shells.Most of the explained variation (12.9% of thetotal variation in the living assemblage and 14.4%in the empty shell assemblage) can be ascribed

to the differences between Sphagnum layers(pseudo-F = 9.35 and pseudo-F = 11.10, P < 0.001,respectively). Summer warming accounted for4.2% of the total variation in the living assem-blage and 6.0% in the empty shell assemblage(pseudo-F = 3.71, P = 0.002 in the living assem-blage; pseudo-F = 5.21, P < 0.001 in the empty shellassemblage).

The relationships between the significant factors(sampling depth and summer warming) and theassemblage structure of living amoebae and emptyshells were illustrated by the ordination diagrams(Fig. 2A, B). In both cases, the first axes accountedfor a considerable part of the variation explainedby the factors (93.3% in the living assemblage and94.2% in the empty shell assemblage) and werepositively correlated with sampling depth and neg-atively with summer warming. The living and emptyshell assemblages in the upper layer were charac-terised by Assulina muscorum and Arcella arenaria,whereas Trigonopyxis minuta and Nebela militarismostly occurred in the deeper layers. In both lay-ers, concentrations of living individuals and emptyshells of Arcella catinus, Bullinularia indica and T.minuta increased in response to summer warming.Negative responses to the treatment were charac-teristic for most of the other taxa, mainly Corythiondubium, Euglypha strigosa glabra, and Euglypharotunda.

-0.6 -0.4 -0.2 0.0 0.2 0.4 0.6

-0.3

-0.2

-0.1

0.0

0.1

0.2

0.3

RDA1 (93.4%)

RD

A2

(6.6

%) Assmus

ArcareArccat

Bulind

CordubEugstr

EugstgEugrot

Nebmil

NebtinTriarc

Trimin

Upper

Deeper

Control

Warmed

A

-0.2 0.0 0.2 0.4

-0.2

-0.1

0.0

0.1

0.2

RDA1 (94.2%)

RD

A2

(5.8

%) Assmus

Arcare

Arccat

Archeu

Arlfla

Bulind

Cordub

Cyceur

Eugstr

EugstgEugrot

Nebmil

Nebtin

TriminrT ilin

Upper

Deeper

Control

Warmed

B

Figure 2. Ordination diagrams (partial RDA) illustrating the significant effects of the sampling depths (upper anddeeper layers) and summer warming (warmed and control) on assemblage structure of living testate amoebae(A) and empty shells (B) after eight years of exposure to the treatments. The ordinations were performedon the matrixes of the Hellinger-transformed species concentrations controlling for the effects of the blocks.The numbers in the parentheses are the percentage of the explained variation in species abundance datarepresented by each axis of the ordination diagram. Biplot scores are represented following type 2 scaling(correlation biplot). Arrows indicate species vectors, black diamonds indicate factor levels. Only species withinter-set correlations ≥ |0.1| with at least one of the axes are shown. Species codes are listed in Appendix A.

406 A.N. Tsyganov et al.

Discussion

Assemblage Composition andEnvironmental Conditions

The testate amoeba assemblage at the experimen-tal site was relatively diverse and was dominated byubiquitous and xerophilous species, yet also char-acterized by frequent occurrence of hydrophilousspecies. The dominant species Assulina musco-rum is considered one of the most common andabundant species in Sphagnum (Beyens et al.1986a; Tolonen 1986). This species is typical fora wide range of moisture conditions but mostabundant in intermediate to dry locations (Bobrovet al. 1999; Charman et al. 2000). Such a dis-tribution is also characteristic of the subdominantspecies Nebela militaris (Charman et al. 2000).The hydrological preferences of the other abun-dant species, Arcella catinus, are not completelyclear. It appears at variable positions on hydrologi-cal gradients (Charman et al. 2000), although someresearchers refer to it as an indicator of dry habi-tats (Lamentowicz et al. 2007). The subdominantspecies Bullinularia indica, Corythion dubium, Try-gonopyxis minuta are normally xerophilous (Bobrovet al. 1999; Lamentowicz and Mitchell 2005). Inspite of most dominant and subdominant speciesindicated dry conditions in the biotope, severalhydrophilous species were also found. Most ofthem (e.g. Arcella discoides, Heleopera sphagni,Hyalosphenia papilio) were represented by a fewshells only, but Archerella flavum was consistentlyfound in the samples and comprised 2% and 5%of the total counts of living and empty shell assem-blages, respectively. Previous studies also reportedoccasional co-occurrence or even co-dominanceof species with opposite ecological preferences atthe same site (Booth 2008; Charman et al. 2007).Booth (2008) suggested that, at sites with inter-mediate topography, this can be explained by awide range of moisture variability throughout theyear so that species with wet and dry optima maybe active at different times. Thus, the assemblagecomposition of testate amoebae at our experimen-tal site indicates predominantly dry conditions withsome wet spells. Dry conditions may be typical ofsummer when precipitation is low and drainage isgood due to a low permafrost level. Wet condi-tions are presumably characteristic of spring, whensnow is melting and drainage is poor due to frozensoils. Overall, taxonomic composition was typicalfor testate amoeba assemblages in Sphagnum,suggesting that our findings may have bearingbeyond the limits of the studied location.

Effects of the Summer Treatments

In agreement with our first hypothesis, summerwarming reduced species richness, mostly due todisappearance of sparse species, and increasedthe proportion of xerophilous species in the livingassemblage of testate amoebae. This suggests thatthe summer warming effects on assemblage struc-ture of testate amoebae mostly operate throughdecreased soil moisture and even a reductionin volumetric water content from 34 to 26% canpromote development of xerophilous species. Incontrast to our expectations, the abundance of liv-ing amoebae remained constant in spite of thedecreased moisture content. This can be explainedby counteracting effects of higher summer temper-atures which could possibly affect testate amoebaabundance indirectly through increased bacterialbiomass (Beyens et al. 2009; Jassey et al. 2011a).

Summer warming induced similar responses indiversity and structure of the empty shell assem-blages, suggesting that empty shells are goodindicators of corresponding living assemblages.Yet, the total abundance of empty shells alsodeclined under warming in both layers, whereasthat of living individuals was maintained. Probably,empty shells decomposed faster at higher temper-ature, as the alternative explanation of decreasedmortality of living amoebae would not keep thelive abundance constant. Decomposition of emptyshells is a biological process primarily controlled bytemperature and moisture (Coûteaux 1992; Lousierand Parkinson 1981; Meisterfeld and Heisterbaum1986). Most likely, higher temperatures enhancedmicrobial activity, resulting in faster decompositionand accelerated nutrient cycling. This assumptionis supported by findings of our companion study(Dorrepaal et al. 2009), which reported increasedecosystem respiration rates – heterotrophic respi-ration in particular – in response to warming in theOTCs during spring and summer. This supportsthe hypothesis that empty shells can be indica-tors of microbial activity in ecosystems (Meisterfeld1980; Wanner 1991). However, it may be difficult todisentangle the effects of the corresponding livingassemblage and microbial activity on the abun-dance of empty shells.

Effects of the Winter Treatments

Contrary to our second hypothesis, increasedwinter snow precipitation did not affect any charac-teristics of the living and empty shell assemblagesof testate amoebae in either of the Sphagnumlayers. We expected that a thicker snow cover

Seasonal Climate Effects on Testacea 407

would mitigate negative effects of low soil tempera-tures, thereby increasing abundance and diversityof the assemblages. Apparently, the increase of1.0–1.7 ◦C is not crucial for survival of testateamoeba cysts in the range of the sub-zero win-ter soil temperatures at the experimental site. Itis also evident that these temperature changescould not promote activity of testate amoebaebecause of the frozen soil water. Increased win-ter precipitation combined with spring warming onlyreduced abundance of empty shells in both layerswhereas the living assemblages were unaffected.In this case, the decrease in abundance of emptyshells may again relate to the increased decom-position rates in response to warming (Dorrepaalet al. 2009). The lack of effects of the treatmenton the living assemblages was unexpected as wehypothesised that melting of the increased snowcover and higher spring temperatures would pos-itively affect abundance and species diversity andwould cause shifts in assemblage structure towardshydrophilous species. The neutral effects can beexplained by the unaffected soil moisture contentat the experimental site in spring. Along with theresponses of the living assemblages to summerwarming, these data support the hypothesis aboutthe key role of substrate moisture for testate amoe-bae.

Seasonal Interactions and SamplingDepth Effects

In contrast to our expectations, the effects of theseasonal climate manipulations on testate amoe-bae were additive and consistent in both living anddecaying layers of Sphagnum. The additive effectsof the treatments can relate to the weak impactof the snow addition on testate amoebae, so thedetected changes in the assemblages were drivenby one group of species that were sensitive to sum-mer warming. The consistency in the responsesof testate amoebae with depth could be explainedby the relatively uniform assemblage structure oftestate amoebae along the Sphagnum profile. Ver-tical stratification of testate amoebae in Sphagnumresults from complex interactions among verticalgradients of light, temperature, moisture, concen-trations of building particles for shells, decayingorganic matter and the downwards transfer of shells(Booth 2002; Heal 1962, 1964; Jassey et al. 2011b;Mazei et al. 2007; Meisterfeld 1977, 1978; Mieczan2010; Mitchell and Gilbert 2004; Schönborn 1963).It has also been shown, that the patterns of verti-cal stratification can vary depending on micrositeconditions being less pronounced in dry biotopes

(Booth 2002). At the studied site, only sparsespecies were restricted in their distribution to onelayer only. The dominant and subdominant speciesdid occur in both layers nevertheless showing well-defined variation in the relative abundance withdepth. In terms of the assemblage characteristics,empty shells showed a clear stratification with depthbeing more abundant and diverse in the deeperlayer. However, these patterns can be related toaccumulation of empty shells in the deeper layerdue to Sphagnum growth and downwards trans-fer of shells. This accumulation can also explainthe consistency in responses of empty shells withdepth as they can integrate the effects of the climatemanipulations over the duration of the experiment.The living assemblage was less heterogeneous interms of assemblage level characteristics and onlygreater evenness contributed to the greater speciesdiversity in the deeper layer. Thus, in dry biotopes,testate amoeba assemblages in living and decayinglayers of Sphagnum are equally sensitive to climatechange during the growing season.

We conclude that both living and empty shellassemblages of Sphagnum-dwelling testate amoe-bae are responsive to climate-induced changesin high-latitude bogs. In a subarctic climate, soilwarming in the growing season has stronger effectson the assemblages than increased soil tempera-tures associated with higher snow cover in winter.Assemblage structure and species diversity of bothliving amoebae and empty shells are most sensi-tive to the climate-induced changes. The increasedproportion of xerophilous species suggests thatthe effects may relate to decreased soil moisture.Higher soil temperatures during the growing sea-son can also decrease abundance of empty shellspresumably due to increased decomposition rates.The responses of testate amoeba assemblagesto climate changes are consistent with depth indry biotopes with relatively homogeneous verti-cal distribution of testate amoebae. Thus, althoughpossible effects of climate changes on preserva-tion of empty shells should always be taken intoaccount, species diversity and structure of testateamoeba assemblages in dry subarctic bogs aresensitive proxies for climatic changes during thegrowing season.

Methods

Study site: The study was performed on a north-facing slopingbog near the Abisko Scientific Research Station in northernSweden (68◦21′N, 18◦49′E, 340 m a.s.l.). At this site, theannual precipitation amounts to 320 mm year–1 with a mean

408 A.N. Tsyganov et al.

Table 2. Experimental design and treatment codesused for the climate manipulations.

Summer treatment

Winter treatment Ambient Warming

Ambient AA WASnow addition AS WSSnow addition plus

spring warmingAS+ WS+

summer temperature of 7 ◦C and a mean winter tempera-ture of −6 ◦C. The length of the growing season is 130 days(Karlsson and Callaghan 1996). The moss component of thebog is dominated by Sphagnum fuscum (Schimp.) H. Klinggr.Other bryophytes and lichens are sparse. The vascular plantcommunity is low and open (maximum shrub height ∼15 cm,average cover ∼20%), and consists mainly of Empetrum nigrumssp. hermaphroditum (Hagerup) Böcher; Rubus chamaemorusL.; Andromeda polifolia L.; Vaccinium microcarpum (Turcz. exRupr.) Schmalh.; Betula nana L.; Vaccinium uliginosum L.; andCalamagrostis lapponica (Wahlenb.) Hartm. The permafrost ispresent at a depth of approximately 55 cm (Dorrepaal et al.2009). The site is relatively dry and a true water table on the topof the frozen peat layer is rare and below 30 cm depth duringmost of the summer (Dorrepaal et al. 2009).

Experimental design: The experiment represents six in situclimate manipulations consisting of full factorial combinationsof two summer treatments (ambient, warming) and three wintertreatments (ambient, snow addition, snow addition plus springwarming), as was described in detail by Dorrepaal et al. (2004).Briefly, the six combinations of the treatments (Table 2) hadfive replicated plots arranged in a randomized block design.All of the manipulations were performed by using a modifiedlarger version (2.2 m in diameter) of the open-top cham-bers (OTCs) proposed by the International Tundra ExperimentProject (Marion et al. 1997). The summer warming treatments(W) involved OTC presence from early June until late Septem-ber. For the winter manipulations, the OTCs served as passivesnow traps on the snow addition (S) and snow addition plusspring warming (S+) plots from late September until late April.From late April through May, after virtually all snow had melted,the OTCs were removed from half of the snow addition treat-ments (S) while the other half kept their OTC for spring warmingplots (S+).

The climate manipulations had significant and consistenteffects on soil temperature and snow depth (Dorrepaal et al.2009). The presence of OTCs increased mean daily soil tem-peratures in the top 5 cm of Sphagnum in June and July, but notin August and September. The soil temperatures in the summerwarming plots were about 1.2 ◦C higher than the ambient meanof 10.2 ◦C in the control plots (here and further, the results ofcontinuous temperature records in 2002 – 2004). The snowaddition manipulation doubled the snow cover and increaseddaily mean soil temperatures by 0.7 – 1.4 ◦C above the ambi-ent mean of –4.3 ◦C. Absolute minimum temperatures were onaverage approximately 6 ◦C higher in the snow addition plotsthan in the control. The OTCs did not affect the duration of snowcover in winter. The spring treatment elevated daily mean soiltemperature by 1.0 – 1.1 ◦C above the 3.4 ◦C in the control. Thesummer warming treatment reduced the mean volumetric soilwater content from 34% down to 26% but only after five years(Dorrepaal et al. 2009). Maximum active layer depth was not

affected by any of the climate manipulations. Analyses of thetreatment effects on the vegetation can be found in publicationsof Aerts et al. (2004, 2006, 2009), Dorrepaal et al. (2004, 2006,2009) and Keuper et al. (2011).

Sampling strategy: Testate amoebae were sampled aftereight years of the climate manipulations on August 23, 2008.In each plot, a 14 cm deep Sphagnum sample was collectedby pushing a cork corer (2.0 cm diameter, 14.0 cm length) intothe moss until it was level with the moss surface. Each samplewas divided into two parts according to the vertical stratificationof the Sphagnum sward. The top parts included capitulum andliving stems, had a length of 1.0–2.5 cm and represented theupper Sphagnum layer. The rest of the sample below the livingparts contained decaying Sphagnum peat and represented thedeeper layer. All 60 samples were stored in screw-top plasticvials and fixed with 3% neutralised formaldehyde.

Laboratory methods: Sample preparation for counting oftestate amoebae followed a slightly modified version of thewater-based procedure (Hendon and Charman 1997). First,exotic marker spores (five tablets of Lycopodium spores; LundUniversity Batch № 483216) were added to each sample forquantitative analysis (Stockmarr 1971). After that, the sampleswere thoroughly shaken for 10 min to extract the shells fromSphagnum stems. The suspension was sequentially passedthrough sieves with mesh openings of 300 and 10 �m, andthe fraction between the two sieves was retained. We used afiner sieve mesh for back sieving instead of the 15 �m meshsuggested by the original technique in order to reduce the pos-sible loss of small species (Payne 2009). The moss materialwhich remained on the 300-�m sieve was oven dried at 60 ◦Cfor 48 hours and Sphagnum dry weight was determined. RoseBengal was added to the samples in order to distinguish emptyshells from living organisms. Testate amoebae were identifiedand counted in a drop of the concentrate mixed with glycerol atmagnification ×400 using a light microscope (Olympus BX-50).Living individuals and empty shells were counted separately.Encysted testate amoebae were not numerous (less than 2%)and were considered as living organisms. A minimum total of200 shells was counted in each sample.

Numerical analyses: The effect of the climate manipula-tions on the living and empty shell assemblages of testateamoebae in the living and decaying Sphagnum layers wastested according to a linear mixed-effects model which includedfour factors: (1) block (five levels, random), (2) summer treat-ments (two levels, fixed), (3) winter-cum-spring treatments(three levels, fixed), (4) sampling depth (two levels, fixed,orthogonal to the blocks and treatments), with five replicationsper combination of factor levels. Blocks were included amongthe factors in order to reduce the variability of the data andincrease the power of statistical tests. The model was tested formain effects of the fixed factors and interactions among them.Significant winter treatment effects were followed by plannedorthogonal comparisons in order to test the null hypotheseswhether the snow addition or the snow addition plus springwarming plots differed from the control. Statistical tests wereconsidered significant at P < 0.05.

Analysis of variance (ANOVA) was performed for testingeffects of the model factors on univariate characteristics ofthe testate amoeba assemblages. Both living and empty shellassemblages were assessed for: (1) concentrations (*103 shellsg-1 of Sphagnum dry weight), (2) species richness (number ofspecies per sample), (3) Shannon-Wiener’s diversity index, (4)Pielou’s evenness index. The assumptions of parametric testswere estimated by visual inspection of probability plots of resid-uals and plots of residuals versus predicted values. A naturallogarithm transformation [y′ = loge (y)] was applied to all uni-

Seasonal Climate Effects on Testacea 409

variate characteristics of the testate amoeba assemblages asit improved normality and homogeneity of variance among thetreatments.

Partial redundancy analysis (partial RDA) was performed inorder to examine the relationships between the model factorsand the assemblage structure of testate amoebae. Partial RDAis a constrained ordination technique which is particularly suit-able for analysis of multivariate response data because it allowsfor assessment of the amount of variation explained by each ofthe factors and their interactions in multifactorial experimen-tal designs using a pseudo-F ratio and P-values determinedby a non-parametric permutation procedure (Legendre andAnderson 1999). Species concentration data of living testateamoebae and empty shells were analysed separately and wereconsidered as response variables, whereas the factors of themodel were treated as predictors. The model factors werecoded as orthogonal Helmert contrasts to allow testing thefactors and interaction in a way that provides the correct Fvalues (Borcard et al. 2011). Prior to the analysis, speciesencountered in two samples or less (three species in the livingassemblage: Difflugia pristis, Difflugia manicata, Hyalospheniaelegans, and six species in the empty shell assemblage: Arcellaarenaria sphagnicola, Difflugiella oviformis, Hyalosphenia ele-gans, Hyalosphenia papilio, Nebela (Argynnia) dentistoma,Nebela flabellulum) were excluded from the datasets in order toeliminate the unwarranted effect of rare taxa on the ordinationresults. The maximal relative abundance of the removed taxadid not exceed 6.1% per sample in the empty shell assemblageand 5.1% per sample in the living assemblage. The speciesconcentrations were Hellinger-transformed so that the Hellingerdistance, which is a more appropriate dissimilarity measurefor community composition data than Euclidian distance, waspreserved in RDA based on the covariance matrix (Legendreand Gallagher 2001). Effects of each of the fixed factors and

their interactions on the testate amoeba assemblages wereassessed by partialling out the effects of all the other factors inthe model (Legendre and Anderson 1999). Statistical tests ofsignificance were carried out by 999 Monte Carlo permutationsrestricted within block under the reduced model. Unadjusted R2

values were corrected following Peres-Neto et al. (2006). Ordi-nation diagrams based on the results of RDA were used in orderto visualise the relationships between testate amoeba taxa andsignificant factors.

The statistical analyses were performed in the R-languageenvironment (R Development Core Team 2010). The func-tions ‘decostand’, ‘rda’ and ‘RsquareAdj’ in the package ‘vegan’(Oksanen et al. 2010) were used for the Hellinger transforma-tion, partial-RDA, and R2 values correction, respectively.

Acknowledgements

The fieldwork at the Abisko Scientific Research Sta-tion was financially supported by grants of the RoyalSwedish Academy of Science. The OTC experi-ment was co-funded by the Dutch Polar Program(ALW-NPP grant 851.30.023) and the EU-ATANS(grant FP6 506004). The first author thanks FridaKeuper and Ann Milbau for assistance in the fieldwork.

Appendix A.

See Table A.1.

410

A.N

. T

syganov et

al.Table A.1. Occurrence and mean concentrations of living testate amoebae and empty shells (*103 shells g-1 Sphagnum dry weight) in the living(upper) and decaying (deeper) Sphagnum layers of the experimental plots (n = 5 per combination of the treatments) after eight years of thesummer and winter climate manipulations. Species codes used in the figures are also indicated. Species are ordered in descending order of theiroccurrence. Occurrence is the percentage of the total number of samples in which the taxon was present. Species marked with (*) were foundonly in the empty shell assemblage, with (U) found only in the upper layer, and with (D) found only in the deeper layer. Species concentrationsgreater than 5% of the total concentration in a combination of the treatments are shown in bold. For the treatment codes, see Table 2.

Species Codes Living layer Decaying layer

AA AS AS+ WA WS WS+ AA AS AS+ WA WS WS+

Living assemblageCommon species (occurrence 80 – 100%)

Assulina muscorum Assmus 5.60 5.55 5.42 5.42 7.00 3.00 2.17 2.07 0.70 1.55 1.31 1.70Arcella catinus Arccat 1.19 1.31 3.05 1.11 3.23 2.07 0.12 0.44 0.29 0.99 1.60 1.01Trigonopyxis minuta Trimin 0.45 1.35 0.15 1.08 0.40 0.10 0.80 7.17 1.17 4.27 1.61 0.47

Common species (occurrence 50 – 80%)Bullinularia indica Bulind 0.61 0.24 1.78 0.71 3.85 1.98 0.37 0.22 0.12 0.42 2.22 0.56Corythion dubium Cordub 5.36 0.17 0.27 0.19 0.97 0.17 2.78 1.74 0.84 0.20 1.07 0.34Assulina seminulum Asssem 0.46 0.44 0.32 0.45 0.45 0.39 0.98 0.49 0.45 0.06 0.46 0.33Nebela militaris Nebmil 2.11 0.03 0.06 0.02 0.04 6.97 2.33 1.98 1.19 0.54 0.67Nebela tincta Nebtin 5.05 0.77 0.27 0.27 0.13 0.03 1.34 0.46 0.67 0.26 1.20 0.17

Common species (occurrence 10 – 50%)Trigonopyxis arcula Triarc 0.11 0.14 0.10 0.08 0.03 0.95 1.95 0.88 0.19 0.55 0.06Euglypha strigosa glabra Eugstg 1.01 0.12 0.06 0.04 0.03 1.77 0.89 1.31 0.20 0.06 0.27Hyalosphenia sp1 Hyasp1 0.63 0.33 0.10 0.03 0.19 0.07 0.04 0.12 0.10 0.24Euglypha strigosa Eugstr 0.56 0.07 0.24 0.03 0.03 0.40 1.16 0.13 0.38 0.21Archerella flavum Arlfla 3.80 0.38 0.24 0.02 0.31 0.46 0.01 0.10 0.36 0.03Valkanovia elegans Valele 0.19 0.08 0.14 0.04 0.45 0.23 0.15 0.21Arcella arenaria U Arcare 0.07 0.05 0.06 0.09 0.77 0.70Cyclopyxis eurystoma Cyceur 0.11 0.04 0.02 0.09 1.89 0.35 0.16 0.06 0.03Arcella hemisphaerica undulata Archeu 0.01 0.03 0.11 0.15 0.14 0.07 0.04Euglypha rotunda Eugrot 0.22 0.08 0.02 1.18 0.51 0.15Placocista sp1 Plasp1 1.33 0.04 0.07 0.46 0.79 0.30 0.45Euglypha laevis Euglae 1.34 0.06 0.06 0.33 0.16 0.22Trinema complanatum D Tricom 1.42 0.34 0.33

Rare species (occurrence < 10%)Euglypha ciliata Eugcil 0.01 0.06 0.31 0.34Euglypha ciliata glabra Eugcig 0.12 0.02 0.16 0.34Euglypha filifera spinosa Eugfis 0.03 0.47 0.07 0.12Trinema lineare Trilin 0.04 0.01 0.40Arcella arenaria sphagnicola U Arcars 0.02 0.14Arcella discoides Arcdis 0.01 0.03 0.12Heleopera sphagni Helsph 0.52 0.21

Seasonal

Clim

ate E

ffects on

Testacea

411Table A.1 (Continued)

Species Codes Living layer Decaying layer

AA AS AS+ WA WS WS+ AA AS AS+ WA WS WS+

Difflugia pristis D Difpri 0.33 0.23Difflugia manicata D Difman 0.12Hyalosphenia elegans U Hyaele 0.22

Total number of species 24 21 22 20 16 12 21 20 20 16 18 15Empty shell assemblageCommon species (occurrence 80 – 100%)

Assulina muscorum Assmus 22.97 12.90 10.17 10.11 19.57 3.72 60.74 55.42 48.11 38.42 33.77 12.49Arcella catinus Arccat 2.34 2.01 2.32 2.02 1.95 1.78 5.08 5.34 5.03 5.91 5.00 5.59Corythion dubium Cordub 3.39 0.41 1.38 0.32 0.97 0.44 20.70 12.90 10.82 2.90 5.40 1.44Assulina seminulum Asssem 1.64 0.44 0.38 0.28 0.77 0.27 7.15 4.20 1.08 0.44 3.00 0.55

Common species (occurrence 50 – 80%)Trigonopyxis minuta Trimin 0.41 0.88 0.11 0.67 0.11 0.03 4.06 10.70 1.23 7.10 3.04 0.96Nebela tincta Nebtin 4.00 0.58 0.32 0.26 0.37 0.10 15.59 8.08 7.24 11.45 4.37 0.66Archerella flavum Arlfla 10.59 0.56 0.68 0.02 2.39 14.75 7.24 7.37 2.77 5.50 0.90Euglypha strigosa glabra Eugstg 1.57 0.19 0.02 0.04 0.03 0.04 19.88 10.25 4.34 6.34 1.24 0.51Bullinularia indica Bulind 0.34 0.17 1.01 0.17 0.68 0.29 0.33 0.44 0.12 0.67 1.19 0.60Nebela militaris Nebmil 2.73 0.08 0.11 0.30 0.05 44.04 32.94 11.95 19.72 16.19 0.77Euglypha strigosa Eugstr 1.08 0.14 0.20 0.02 0.04 0.05 2.17 2.21 5.44 1.92 1.25 0.41Hyalosphenia sp1 Hyasp1 0.51 0.23 0.17 0.06 0.10 0.02 0.31 0.12 0.29 0.85 0.79 0.42

Common species (occurrence 10 – 50%)Valkanovia elegans Valele 0.27 0.15 0.46 0.16 0.02 0.74 2.96 2.28 0.68 2.90 0.53Arcella arenaria Arcare 0.59 0.09 0.19 0.30 0.75 0.44 0.43 0.04 0.02Placocista sp1 Plasp1 2.90 0.04 0.22 0.01 5.79 1.51 1.30 0.40 2.47 0.06Euglypha ciliata glabra Eugcig 0.88 0.05 0.04 0.19 2.06 1.13 1.20 3.04 0.97Trigonopyxis arcula Triarc 0.23 0.04 0.21 0.13 0.94 2.48 1.49 1.25 0.67Cyclopyxis eurystoma Cyceur 0.01 0.01 0.02 0.03 5.42 2.53 0.81 0.41 0.79 0.21Euglypha laevis Euglae 1.15 0.05 0.66 0.20 0.13 0.82 0.84 3.83 4.73 2.78 0.27Arcella hemisphaerica undulata Archeu 0.06 0.07 0.09 0.19 0.16 0.05 0.02 0.06 0.13 0.05Euglypha rotunda Eugrot 0.33 0.06 0.06 2.87 1.64 1.90 0.74 0.29 0.27Trinema lineare D Trilin 1.38 0.49 0.42 0.17 0.30 0.22Euglypha ciliata Eugcil 0.02 0.03 0.16 0.09 2.97 0.34 1.12Euglypha filifera spinosa Eugfis 0.03 0.78 0.15 0.77Heleopera sphagni D Helsph 1.09 0.12 0.54Trinema complanatum D Tricom 1.26 0.04 0.34 0.28

412 A.N. Tsyganov et al.

Rar

e

spec

ies

(occ

urre

nce

<

10%

)D

ifflug

ia

pris

tisD

Difp

ri

3.17

1.80

0.22

Arc

ella

aren

aria

spha

gnic

ola

UA

rcar

s0.

11

0.08

Arc

ella

disc

oide

sA

rcdi

s0.

01

0.05

0.12

Diffl

ugia

man

icat

aD

Difm

an

2.39

0.81

Arc

ella

gibb

osa*

DA

rcgi

b0.

09

0.12

Hya

losp

heni

a

eleg

ans

Hya

ele

0.11

2.14

Neb

ela

(Arg

ynni

a)

dent

isto

ma*

DN

ebde

n

0.31

Diffl

ugie

lla

ovifo

rmis

*DD

ilovi

0.16

Hya

losp

heni

a

papi

lio*D

Hya

pap

0.16

Neb

ela

flabe

llulu

m*D

Neb

fla

0.82

Tota

l num

ber

of

spec

ies

23

21

19

20

18

18

32

26

26

24

25

20 References

Aerts R, Cornelissen JHC, Dorrepaal E (2006) Plant perfor-mance in a warmer world: General responses of plants fromcold, northern biomes and the importance of winter and springevents. Plant Ecol 182:65–77

Aerts R, Callaghan TV, Dorrepaal E, van Logtestijn RSP,Cornelissen JHC (2009) Seasonal climate manipulationsresult in species-specific changes in leaf nutrient levels andisotopic composition in a sub-arctic bog. Funct Ecol 23:680–688

Aerts R, Cornelissen JHC, Dorrepaal E, van Logtestijn RSP,Callaghan TV (2004) Effects of experimentally imposed cli-mate scenarios on flowering phenology and flower productionof subarctic bog species. Glob Change Biol 10:1599–1609

Beveridge OS, Humphries S, Petchey OL (2010) The inter-acting effects of temperature and food chain length on trophicabundance and ecosystem function. J Anim Ecol 79:693–700

Beyens L, Ledeganck P, Graae BJ, Nijs I (2009) Are soil biotabuffered against climatic extremes? An experimental test ontestate amoebae in arctic tundra (Qeqertarsuaq, West Green-land). Polar Biol 32:453–462

Beyens L, Chardez D, Delandtsheer R, Debock P, JacquesE (1986) Testate amebas populations from moss and lichenhabitats in the Arctic. Polar Biol 5:165–173

Bobrov AA, Charman DJ, Warner BG (1999) Ecology of tes-tate amoebae (Protozoa: Rhizopoda) on peatlands in westernRussia with special attention to niche separation in closelyrelated taxa. Protist 150:125–136

Booth RK (2002) Testate amoebae as paleoindicators ofsurface-moisture changes on Michigan peatlands: modernecology and hydrological calibration. J Paleolimnol 28:329–348

Booth RK (2008) Testate amoebae as proxies for mean annualwaiter-table depth in Sphagnum-dominated peatlands of NorthAmerica. J Quat Sci 23:43–57

Borcard D, Gillet F, Legendre P (2011) Numerical Ecologywith R. Springer, New York

Charman DJ (2001) Biostratigraphic and palaeoenvironmentalapplications of testate amoebae. Quat Sci Rev 20:1753–1764

Charman DJ, Blundell A, Members A (2007) A new Europeantestate amoebae transfer function for palaeohydrological recon-struction on ombrotrophic peatlands. J Quat Sci 22:209–221

Charman DJ, Hendon D, Woodland WA. (2000) The Identi-fication of Testate Amoebae (Protozoa: Rhizopoda) in Peats.QRA Technical Guide No 9. Quaternary Research Assosiation,London.

Christensen JH, Hewitson B, Busuioc A, Chen A, Gao X,Held R, Jones R, Kolli RK, Kwon WK, Laprise R, MaganaRueda V, Mearns L, Menéndez CG, Räisänen J, Rinke A,Sarr A, Whetton P, et al. (2007) Regional Climate Projections.In Solomon S, Qin D, Manning M, Chen Z, Marquis M, AverytKB, Tignor M, Miller HL (eds) Climate Change 2007: The Physi-cal Science Basis. Contribution of Working Group I to the FourthAssessment Report of the Intergovernmental Panel on ClimateChange. Cambridge University Press, Cambridge, New York,pp 847–940

Coûteaux MM (1992) Decomposition of cells and empty shellsof testate ameobae (Rhizopoda, Testacea) in an organic-acid

Seasonal Climate Effects on Testacea 413

soil sterilized by propylene-oxide fumigation, autoclaving, andgamma-ray irradiation. Biol Fertil Soils 12:290–294

Dorrepaal E, Aerts R, Cornelissen JHC, Callaghan TV, vanLogtestijn RSP (2004) Summer warming and increased win-ter snow cover affect Sphagnum fuscum growth, structure andproduction in a sub-arctic bog. Glob Change Biol 10:93–104

Dorrepaal E, Aerts R, Cornelissen JHC, Van Logtestijn RSP,Callaghan TV (2006) Sphagnum modifies climate-changeimpacts on subarctic vascular bog plants. Funct Ecol 20:31–41

Dorrepaal E, Toet S, van Logtestijn RSP, Swart E, van deWeg MJ, Callaghan TV, Aerts R (2009) Carbon respirationfrom subsurface peat accelerated by climate warming in thesubarctic. Nature 460:616–679

Gilbert D, Amblard C, Bourdier G, Francez AJ (1998) Themicrobial loop at the surface of a peatland: Structure, function,and impact of nutrient input. Microb Ecol 35:83–93

Heal OW (1962) Abundance and micro-distribution of testateamoebae (Rhizopoda-Testacea) in Sphagnum. Oikos 13:35–47

Heal OW (1964) Observations on the seasonal and spatial dis-tribution of Testacea (Protozoa, Rhizopoda) in Sphagnum. JAnim Ecol 33:395–412

Hendon D, Charman DJ (1997) The preparation of testateamoebae (Protozoa: Rhizopoda) samples from peat. Holocene7:199–205

Ives AR (1995) Predicting the response of populations to envi-ronmental change. Ecology 76:926–941

Jassey VEJ, Chiapusio G, Mitchell EAD, Binet P, ToussaintML, Gilbert D (2011a) Fine-scale horizontal and vertical micro-distribution patterns of testate amoebae along a narrow fen/boggradient. Microb Ecol 61:374–385

Jassey VEJ, Gilbert D, Binet P, Toussaint ML, Chiapusio G(2011b) Effect of a temperature gradient on Sphagnum fallaxand its associated living microbial communities: a study undercontrolled conditions. Can J Microbiol 57:226–235

Jiang L, Morin PJ (2004) Temperature-dependent interactionsexplain unexpected responses to environmental warming incommunities of competitors. J Anim Ecol 73:569–576

Karlsson PS, Callaghan TV (1996) Plant ecology in the sub-arctic Swedish Lapland. Ecol Bull 45:220–227

Keuper F, Dorrepaal E, van Bodegom PM, Aerts R, vanLogtestijn RSP, Callaghan TV, Cornelissen JHC (2011) Arace for space? How Sphagnum fuscum stabilized vegeta-tion composition during long-term climate manipulations. GlobChange Biol 17:2162–2171

Lamentowicz M, Mitchell EAD (2005) The ecology of testateamoebae (Protists) in Sphagnum in North-Western Poland inrelation to peatland ecology. Microb Ecol 50:48–63

Lamentowicz M, Tobolski K, Mitchell EAD (2007) Palaeoeco-logical evidence for anthropogenic acidification of a kettle-holepeatland in northern Poland. Holocene 17:1185–1196

Lamentowicz M, Van der Knaap W, Lamentowicz L, VanLeeuwen JFN, Mitchell EAD, Goslar T, Kamenik C (2010) Anear-annual palaeohydrological study based on testate amoe-bae from a sub-alpine mire: surface wetness and the role ofclimate during the instrumental period. J Quat Sci 25:190–202

Legendre P, Anderson MJ (1999) Distance-based redundancyanalysis: Testing multispecies responses in multifactorial eco-logical experiments. Ecol Monogr 69:1–24

Legendre P, Gallagher ED (2001) Ecologically meaning-ful transformations for ordination of species data. Oecologia129:271–280

Lousier JD (1974a) Effects of experimental soil-moisture fluc-tuations on turnover rates of Testacea. Soil Biol Biochem6:19–26

Lousier JD (1974b) Response of soil Testacea to soil-moisturefluctuations. Soil Biol Biochem 6:235–239

Lousier JD, Parkinson D (1981) The disappearance of theempty tests of litter- and soil-testate amoebae (Testacea, Rhi-zopoda, Protozoa). Arch Protistenkd 124:312–336

Marion GM, Henry GHR, Freckman DW, Johnstone J, JonesG, Jones MH, Levesque E, Molau U, Molgaard P, Par-sons AN, Svoboda J, Virginia RA (1997) Open-top designsfor manipulating field temperature in high-latitude ecosystems.Glob Change Biol 3:20–32

Maxwell B (1992) Arctic Climate: Potential for Change underGlobal Warming. In Chapin FS, Jefferies RL, Reynolds JF,Shaver GR, Svoboda J (eds) Arctic Ecosystems in a Chang-ing Climate: An Ecophysiological Perspective. Academic Press,Inc, San Diego, pp 11–34

Mazei YA, Tsyganov AN, Bubnova OA (2007) The speciescomposition, distribution, and structure of a testate amoebacommunity from a moss bog in the middle Volga river basin.Zool Zh 86:1155–1167

McCarty JP (2001) Ecological consequences of recent climatechange. Conserv Biol 15:320–331

Meisterfeld R (1977) Horizontal and vertical distribution oftestacea (Rhizopoda-Testacea) in Sphagnum. Arch Hydrobiol79:319–356

Meisterfeld R (1978) Die Struktur von Testaceenzönosen(Rhizopoda, Testacea) in Sphagnum unter besonderer Berück-sichtigung ihrer Diversität. Verh Ges Oekol 7:441–450

Meisterfeld R (1980) Die Struktur von Testaceenzönosen (Rhi-zopoda,Testacea) in Böden des Sollings. Verh Ges Oekol8:435–447

Meisterfeld R, Heisterbaum M (1986) The decay of emptytests of testate amoebae. Symp Biol Hung 33:285–290

Mieczan T (2010) Vertical microzonation of testate amoebaeand ciliates in peatbog waters in relation to physical and chem-ical parameters. Pol J Ecol 58:729–739

Mitchell EAD, Gilbert D (2004) Vertical micro-distribution andresponse to nitrogen deposition of testate amoebae in Sphag-num. J Eukaryot Microbiol 51:480–490

Mitchell EAD, Buttler AJ, Warner BG, Gobat JM (1999) Ecol-ogy of testate amoebae (Protozoa: Rhizopoda) in Sphagnumpeatlands in the Jura mountains, Switzerland and France. Eco-science 6:565–576

Mitchell EAD, Charman DJ, Warner BG (2008) Testateamoebae analysis in ecological and paleoecological stud-ies of wetlands: past, present and future. Biodivers Conserv17:2115–2137

414 A.N. Tsyganov et al.

Moore PD, Bellamy DJ (1974) Peatlands. Springer, New York

Norf H, Arndt H, Weitere M (2007) Impact of local temperatureincrease on the early development of biofilm-associated ciliatecommunities. Oecologia 151:341–350

Oksanen J, Blanchet FG, Kindt R, Legendre P, O’Hara RB,Simpson GL, Solymos P, Stevens MHH, Wagner H. (2010)Vegan: Community Ecology Package. R package version 1.17-2. http://CRAN.R-project.org/package=vegan.

Payne R (2009) The standard preparation method for testateamoebae leads to selective loss of the smallest taxa. QuatNewslett 119:16–20

Peres-Neto PR, Legendre P, Dray S, Borcard D (2006)Variation partitioning of species data matrices: Estima-tion and comparison of fractions. Ecology 87:2614–2625

Petchey OL, McPhearson PT, Casey TM, Morin PJ (1999)Environmental warming alters food-web structure and ecosys-tem function. Nature 402:69–72

R Development Core Team (2010) R: A Language and Envi-ronment for Statistical Computing. R Foundation for StatisticalComputing, Vienna. http://www.R-project.org.

Savage VM, Gillooly JF, Brown JH, West GB, Charnov EL(2004) Effects of body size and temperature on populationgrowth. Am Nat 163:429–441

Schönborn W (1963) Die Stratigraphie lebender Testaceen imSphagnetum der Hochmoore. Limnologica 1:315–321

Shaver GR, Canadell J, Chapin FS, Gurevitch J, Harte J,Henry G, Ineson P, Jonasson S, Melillo J, Pitelka L, Rus-tad L (2000) Global warming and terrestrial ecosystems: Aconceptual framework for analysis. Bioscience 50:871–882

Stockmarr J (1971) Tablets with spores used in absolute pollenanalysis. Pollen Spores 13:615–621

Tolonen K (1986) Rhizopod Analysis. In Berglund BE (ed)Handbook of Holocene Paleoecology and Paleohydrology. JohnWiley & Sons, Chichester, pp 467–468

Traill LW, Lim MLM, Sodhi NS, Bradshaw CJA (2010)Mechanisms driving change: altered species interactions andecosystem function through global warming. J Anim Ecol79:937–947

Tsyganov AN, Nijs I, Beyens L (2011) Does climate warmingstimulate or inhibit soil protist communities? A test on tes-tate amoebae in High-Arctic tundra with Free-Air TemperatureIncrease. Protist 162:237–248

Walther GR, Post E, Convey P, Menzel A, Parmesan C,Beebee TJC, Fromentin JM, Hoegh-Guldberg O, Bairlein F(2002) Ecological responses to recent climate change. Nature416:389–395

Wanner M (1991) Zur Morphologie von Thekamoben (Proto-zoa: Rhizopoda) in süddeutschen Wäldern. Arch Protistenkd140:236–288

Wilkinson DM, Mitchell EAD (2010) Testate amoebae andnutrient cycling with particular reference to soils. GeomicrobiolJ 27:520–533